Hi,

I guess that I am a typical user that ColorMunki is intended for. Interested in quality and possibly also in color science but not willing to invest thousands of dollars into equipment. Now I had my ColorMunki for a week, time to sum up my experience.

There have been some negative comments on software. I have using ColorMunki on both Mac OS X.4 and Windows XP. I have no complaints on software on Mac OS X.4. On XP it seems at it more cumbersome. There may be some glitches on Mac OS X, for some reason the problems on XP seem to be the same but squared ;-)

Now for the positive things:

1) Profiles are very good, at least when analysed with "Gamut Vision" or checked with "Twenty Balls" from Bill Atkinsons site.
2) The software works very well on Mac OS-X.
3) It is very easy to produce good profiles.
4) It is very easy to scan a printout of a "Macbeth" Color Checker. It would probably be as easy to scan a real Color Checker, but I have only two "Mini Color Checkers". Anyway the ability to scan printouts is very helpful in quantifying what you try to achieve.
5) Color Munki gives not just LAB data but also spectral data. I cannot really put it to use, but it gives me confidence that the data is good enough to be able to handle metamerism.

Now for the negatives:

1) I have seen some glitches especially on Windows XP.
2) The software is somewhat dumbed down. Documentation is not good at all.
3) I cannot see any real advantage compared with Print Fix Pro in real printouts. In theory ColorMunki is much better, but in real world there may not be a lot of difference.

Conclusions:

1) If you happen to have a Print Fix Pro it may be good enough.
2) ColorMunki works fine.
3) The software does not really utilize the hardware. Opportunity for third party vendors to build software which fully employs the quality of data coming from the Color Munki.

Best regards
Erik

Thanks for your conclusions...i can now make decisions to go buy this or not...unlike all the other conversations that lead to no decisions...

Hi!

Now that I have the Color Munki a couple more days I found out that Xrite has quite informative FAQ on their web site. I also gained some experience with the "Color Picker". That said I'm still not satisfied with the documentation, even if the FAQ is very helpful.

Best regards
Erik

I just ordered a monkey for myself! hope it plays along well with me.

I decided to try the Munki after all and am glad I did. I purchased the Photo version primarily to profile my Epson printers and my laptop. I am on XP Pro and Vista and have had no glitches.

The laptop calibration and profile caused a dramatic change from the default on this new unit. In my office I still calibrate/profile my CRT's with an old I1 monitor.

I have profiled three papers so far and have just finished evaluating them. The papers used are Harman Gloss FB AI, Hahnemuhle's FA Baryta and Photo Rag Pearl. When the profiles are viewed in ColorThink they are all quite neutral. When compared against the canned profiles using the 3D gamut mapping featue in ColorThink the Munki's. When printed out and compared the difference ranged from very subtle to quite obviously more pleasing in color rendition and shadow detail {:. I compared all three using Perceptual and Relative Colorimetric renderings. The test image used is the PDI_Taget-DCP, the one with the 21 step gray scale on the side.

My conclusion after just just a few days with this is it produces good profiles, is very easy to use and is well built. My hope is for future upgrades to include at least options for luminance and gamma (or native) in the Display module. This my first venture into profiling printers so I cannot comment yet on that module's strengths or weaknesses. The very thin Teflon sliders also tend to stick on some papers while sliding along the target. No regrets, quite happy in fact.

Dave

David,

You said:

"When printed out and compared the difference ranged from very subtle to quite obviously more pleasing in color rendition and shadow detail."

Do the prints MATCH your calibrated display? That's the most important for trusting color edits.

Yes, quite well actually, I just printed two images out. I must state I do not have a viewing booth, I view prints in (indirect) daylight where I hang most of my prints. I softproof and use curves presets as a start for each paper and intent. I didn't have problems matching display and print, I was dissatisfied with the manufacturers profiles in general. I will do more testing (and buy more ink!!!).

There is an option at the end of the printer profiling process to set the AppSetter, I leave it unticked and have deleted the plug-in with no problems.

Hope that answers your question, the A to B tables for softproofing, IMHO, are quite well built.

Dave

What would be a really good test for accuracy would be to apply a pronounced edit like an off color cast combined with a density and color table adjusts using Replace Color or Selective Color and then see if those edits are reproduced faithfully on the print.

Even if the display image looks ugly the print should look just as ugly and match.

I've never tried this technique, but thought I'ld ask to test how good products like this really are at building accurate printer profiles.

Hi,

Problem is that you need good viewing light for that. Colors probably match pretty well and I feel that everyday prints turn out perfect. The problem I see is that dark parts of the image (like shadow detail) is lost, unless you have very good viewing light. Actually i don't see a way around it, if you have good blacks and shadows you need good light. A 20W halogen spot 2-3 meters from the print is OK.

Something I do is that I print a calculated Macbeth Color Checker and read it with the ColorMunki in scan mode. I can't say that I can see much difference (if any) between my Color Checker and my print, but that also depends on viewing light. The reason I do these checks is essentially to have some kind of "objective" way of checking consistency. I have written a small program to calculate Delta E, something like this:

erik-kaffehrs-computer:~/Projects/DeltaE ekr$ ./DeltaE MBCC_DRYCREEK.csv SP3800_IGSG_01_RelCol_cured_a.csv
6.33 5.52 4.08 5.06 7.75 6.46
8.41 6.75 7.08 4.95 7.66 8.5
8.84 6.7 6.83 6.1 10.7 6.86
5.41 3.41 3.57 2.67 1.7 1.73
Max error = 10.7 mean error 5.96

My DeltaE calculation is the the oldest and simplest form of Delta E, but I guess it's OK for my purpose.